| EPSRC Reference: |
GR/L34419/01 |
| Title: |
EVALUATION AND COMPARISON OF HUMAN AND STATISTICAL LANGUAGE MODEL PERFORMANCE USING MISSING-WORD TESTS |
| Principal Investigator: |
Smith, Professor F |
| Other Investigators: |
|
| Researcher Co-Investigators: |
|
| Project Partners: |
|
| Department: |
Computer Science |
| Organisation: |
Queen's University of Belfast |
| Scheme: |
Standard Research (Pre-FEC) |
| Starts: |
15 September 1997 |
Ends: |
14 September 1998 |
Value (£): |
37,766
|
| EPSRC Research Topic Classifications: |
| Human Communication in ICT |
|
|
| EPSRC Industrial Sector Classifications: |
| No relevance to Underpinning Sectors |
|
|
| Related Grants: |
|
| Panel History: |
|
|
Summary on Grant Application Form |
|
Recently we have experimented with using missing-word tests to evaluate statistical language models designed for speech recognition. A major advantage of this approach is that, by having humans complete the same test set, we can make direct comparisons between human and language model performance. To date we have used this comparison only in a promising pilot study. We now wish to use the experience we have gained to mount a substantial experiment which should provide useful insight into the relative strengths and weaknesses of language model performance, particularly in the use of context information. The main aim will be to derive guidelines on how to bring language model performance closer to human performance.
|
|
Key Findings |
|
This information can now be found on Gateway to Research (GtR) http://gtr.rcuk.ac.uk
|
|
Potential use in non-academic contexts |
|
This information can now be found on Gateway to Research (GtR) http://gtr.rcuk.ac.uk
|
|
Impacts |
|
| Description |
This information can now be found on Gateway to Research (GtR) http://gtr.rcuk.ac.uk |
| Summary |
|
| Date Materialised |
|
|
|
|
Sectors submitted by the Researcher |
|
This information can now be found on Gateway to Research (GtR) http://gtr.rcuk.ac.uk
|
| Project URL: |
|
| Further Information: |
|
| Organisation Website: |
http://www.qub.ac.uk |